Big-C-Culture and Little-c-culture Genres: The Effect of Input Flooding on Speaking Accuracy

1. Introduction

Reading comprehension is among the most complicated cognitive processing abilities where the reader is to decode the written symbols, process the text, understand the meaning and then integrate this semantic load with his/her background knowledge. Reading not develops only writing skill but helps also in improving speaking fluency and accuracy and it is generally deemed that the learners, who read more, are doubtlessly to speak better (Akbar, 2014). In fact, numerous language experts, as Krashen (2004), argues the cessation of the 'complex literacy' which is needed in modern societies. To cure the issue through reading, he presents what he calls free voluntary reading. In many parts of the world, the reading ability is a key factor for attaining success in different aspects of life (Rivers, 1981).As Krashen (2004, p. 23) asserts,

Reading is definitely the sole remedy. To become good readers, to develop a good writing style and to acquire a series of adequate vocabularies, or advanced grammar rules, the unique choice is reading to pick. Reading is the only way by which we can make a good system of spelling.

Krashen sees reading also as a help for the learners to comprehend the spoken language. He believes reading with a long-term exposure and repetition can end in establishing the previous known words and structures. Moreover, an incidentally enjoyable learning can be happening while focusing on the story and meaning (Nation, 2008).In spite of this urgent requirement of communication, the learners cannot express themselves or understand the message sufficiently. Reading in English with the appropriate rate and comprehension has always been a crucial skill, but this important activity has been considered as a frustrating, tedious, and burdensome skill by students because they read slowly, do not enjoy the reading, and do not understand it. The reason can be the lack of sufficient cultural or linguistic information; in addition, they may have some inefficacies of reading as a receptive skill and speaking as a productive skill, which are the two major categories of language abilities.

2. Review of Literature

2.1 Input Flooding

Input flooding has been focused by a number of theorists, but it was Stephen Krashen who formalized this position into a theory known as the input hypothesis, which he later called the 'comprehension hypothesis' (Krashen, 2004). The input hypothesis states that we acquire not learn language by comprehending the input that is a little beyond our current level of competence (Krashen & Terrel, 1983).

Krashen and Terrel (1983) point out that an acquisition can move from the input stage, namely, i (level of competence) to a stage i+1 (immediately following stage along some natural order). Krashen's input hypothesis assigns a crucial role to comprehension. The benefits flooding can be mentioned as easy realization with no disruption in flow of communication, integration of form and meaning in instructions and exposing the learners to a text full of target form examples, in anticipation of attracting their attention (Nemati, & Motallebzade, 2013).

To enhance the power of salient input, input flooding can be useful in which the learners are exposed to the frequent constructions and structures which might promote noticing and the frequent production of that structure (Celce-Murcia, Brinton, Snow, 2014). Input flooding (IF) attempts to make specific features of target language input more frequent and salient. In IF, the input a learner receives is saturated with numerous exam­ples of the target structure with the expectation that this artificial increase will aid the learner in noticing and then acquiring the form (Wong, 2005). It is pointed out by Lee and VanPatten (2003) that the instructors using IF to encourage noticing and processing of specific target structures should provide learners with exposure to both written and oral input activities. IF can also be performed with authentic target language texts.

Input flooding involves reading large amounts of easy texts at the 'i-1' level of comprehensible input below the current level of learners' proficiency (Krashen, 1985). In several respects, Krashen's input hypothesis is the exact opposite of the TG-based models of language learning: where they ignore comprehension, Krashen assigns primary significance to it.

To explore the method of vocabulary enrichment in the realm of Input Flooding, Stahl (2003) focused to study the relationship between input flooding and overall language proficiency. The result proved the existence of several significant differences amongst the learners with regard to vocabulary recognition, meaning, synonym and antonym. Lee (2007) has investigated three consecutive effect of input flooding on the development of reading and vocabulary for Taiwanese university non-English majors. Each study applied a different approach, with latter studies framing the methodology in response to the results of the previous year. These results supported two other findings, expanding different subjects in other countries: (1) input flooding could be unified into an EFL curriculum, termed in-class sustained silent reading, at the university level and (2) input flooding and traditional instruction in acquiring English as a foreign language are of the same level of effectiveness and efficiency, however IF is more effective than the traditional instruction when the treatment duration is longer.

       Rashidi and Piran (2011) investigated the effect of Extensive and Intensive Reading on Iranians' EFL learners' vocabulary size and depth with two intermediate and advanced groups. The results of the study showed that both IR and ER have an impact on learners' vocabulary size and depth significantly and that the students' vocabulary knowledge in terms of size and depth had increased. The study by Pashangzade (2013) revealed input flooding positive effect on the learners' reading comprehension as well. Yamashita (2013) has tried to examine the effect of IF on L2 reading attitude. The result showed increases in comfort and intellectual value and a decrease in anxiety, with no effect on practical value.

2.2 The Nature of Speaking Accuracy

When people hear someone speaks, they pay attention to what the speaker sounds like almost automatically. People are judged in terms of the speaker’s personality, attitudes, home region and native/non-native speaker status. People actually use meaningful speech to make an image of themselves by making pause, speed, pith, variation and intonation Speech is organized into short units linked by thematic connections and synthetic connectors. Some descriptors mentioned the grammatical features of the learners' speech at different levels, and some analytic grids include a separate criterion called Accuracy. Although some experts consider the speaking ability as a crucial individual ability, we should keep in mind that speaking is a part of shared social activity (Luoma, 2009).

The standard of speaking is under the question since first, there are different regional languages and variations which make choosing the specific standard difficult, and second, achieving a native-like standard is a tough. As a result, this standard can be taken as communicative effectiveness (Brown &Yule, 1983).

Speaking accuracy is an attractive element because it is easy to notice and it is available and it should be noticed in evaluating the learner's progress. This evaluation is made in terms of the number of progress, the number of structures, complexity of structures, and the number of the mistakes made by learners (Larsen & Freeman, 1991). It is believed that most language learning is usage-based and learning can be expanded by the length of exposure to these structures (Ellis, 2002). Learners follow a sequence of leaning grammar, learning words, lexical morphemes, phrasal structure, word order, and clause relationships (Pienemann, 1998).

In addition, grammatical features should be taken in to account because they are highly noticed in planned speech situations, namely, lectures, presentations, and discussion for which the speaker has rehearsed and prepared himself in advance for this formal situation (Ochs, 2001). As a result, speaking is a meaningful interaction between people which is made of two major parts, spoken grammar and vocabulary.

As it has been revealed in the study by Huang and Naerssen (1987) IF would play an effective role on speaking structural accuracy. Hafiz and Tudor (1990), in addition, found that reading outside class was the most remarkable predictor of oral communicative ability. In this study the second language learners made the utterances alike the target language. Additionally, Rahmany, Zarei, & Gilak (2013) hold that ER can also be effective on speaking motivation. And Mart (2012) in his study examined the relationship between the IF and speaking skill. The finding of his investigation confirmed the existence of the positive relationship since, he claims, vocabulary and grammatical knowledge are considered as two essential elements of language learning which can influence the learner's speaking performance.

3. Research Questions

The study aims at answering the following questions to find out the effect of input flooding on speaking accuracy.

1. To      what extent does the receptive input flooding through the extensive      literature-oriented (Big C Culture genre) reading have an effect on the      advanced level adult female EFL learners’ speaking structural accuracy?
2. To      what extent does the receptive input flooding through the extensive      culture-oriented (little c culture genre) reading have an effect on the      adult female EFL learners' speaking structural accuracy at the advanced      level?

4. Methodology

4.1 Participants

The participants consisted of two groups of 24 Iranian EFL female students out of 30 learners at the advanced level in Azadi English Institute in Tehran with the age range of 17-30. The speaking course was taught approximately for two semesters successively lasted for about four months. This particular class was chosen as the participants in the advanced language course because they generally had a higher English proficiency than students in the other classes based on the placement test applied. Moreover, because the study of literature was challenging, the students were assumed to have already acquired the basic linguistic and literary skills. The intact groups selected via convenience sampling technique in terms of the availability and then were randomly assigned to two experimental groups, one literature and one culture group based on their scores in the New Cutting Edge placement tests (2013) administered by the institute for the advanced level. To assure the homogeneity of the students, Oxford Placement Test (Allen, 1992) including the grammatical points was administrated. In fact, this test used to assure the homogeneity of the participants and to exclude the learners who were not in advanced level; as a result, 48 students were selected out of 60 learners.

4.2 Instrumentation

To answer the research questions in this study, the students' oral structural accuracy was measured in both groups before and after the study by the means of an IELTS speaking test (2013) Thirteen reading texts were chosen by the learners to be exposed to large amount of input as input flooding technique and discussed the assigned stories in the class with their partners.

To reassure the homogeneity of the learners, the Oxford Placement Test (OPT, Allen, 1992) with twenty multiple-choice items and three parts (reading, structure, & functional test) was conducted. The subsections of the test are as follows:

•  Reading comprehension (question numbers 1-4);
•  Vocabulary (question numbers 5-8);
•  Communication (question numbers 9–12);
•  Grammar (question numbers 13–20).

The cut-off point was 15/20 based on the Oxford Placement test rubric. Since the levels of the EFL learners in our context do not reach the standard level of the native speaker, the students with insignificant range of two scores below the pass grade were included in this program as well (13/20) in terms of the researcher's experience as an English teacher for about 20 years and knowledge of Iranian students. Because the researchers intended to shed some light on the effect of the input flooding on the speaking accuracy, the main focus was on the grammatical part of the given test to assess their grammatical competence. Therefore, there were more grammatical items even in the test. The same test was also used in some other studies (e.g., West, 1997; Dekydtspotter, Schwartz, and Sprouse, 2006; Maleki, 2007; Tsimpli and Dimitrakopoulou, 2007; Wistner and Abe, 2009; Bidabadi and Yamat, 2011).

To answer the research questions, the students' oral structural accuracy was measured in both groups before and after the study by means of IELTS speaking test, Cambridge ESOL, (2013), which is made of three parts (introduction, talking about a particular topic and engagement in a discussion) with the reliability of 0.83–0.86 (Bridges & Shaw, 2004). It was used as pre- and posttests to measure the speaking structural accuracy using the related rating rubric. It should be mentioned that there are two kinds of scores in IELTS: whole-band and half-band and since there is a dramatic difference between a learner who gains seven for example, and the other who gains eight as a result, the raters sometimes used the half-band score to be more accurate.

4.3 Materials

The reading materials employed in this study were Penguin 'graded readers' and Oxford Bookworms, written for EFL learners. They provided a model of English language and motivation for learning and practicing English by maintaining the students' interest. Thirteen story books and novels were read by the experimental groups in which six of them were cultural-oriented such as Pride and Prejudice, Christmas Carol,... and the next seven were literature-oriented texts such as Dorian Gray, Love or Money, ET, so on. The main criterion to differentiate between the two genres was the one proposed by Herron and Dubreil (2000) where a distinction is made between Big C Culture and Little c culture. According to them, Big C Culture refers to a country's literature (Literary works, poems/poets, …), art, and music while Little c culture refers to people's everyday life, beliefs, customs, behavior and values. For the Big C culture which in this study is referred to as 'literature genre' some literary short stories were used and for the little c culture referred to as 'culture genre' hereafter in this study, some texts with related topics to people's everyday life, customs, so on,  were used. Some stories were read in two sessions due to their long length such as Pride & Prejudice, Jane Eyre, Gone with the wind, little women, Frankenstein, The Scarlet Letter, and ET. The books were graded readers, which were available in the library of the institute and chosen by the learners themselves in terms of their levels (i.e., advanced), availability, and interest to further improve their enthusiasm in this study. It is crucial to mention that the levels of the materials were lower than the level of the students and they were all simplified based on the Krashen's (1982) i+1 hypothesis (i.e., the level of the provided graded readers were a little lower than their level).

4.4 Procedure

In order to investigate the impact of the extensive reading program on the learners' speaking accuracy, the study was carried out during 10 sessions in each experimental group. The participants took part in the extensive reading class twice a week and each session took one hour and a half. Both groups were exposed to a two-month extensive reading program with graded readers (mainly simplified novels). The participants were randomly put in two experimental groups. They were all native Persian speakers, who participated in the speaking course of the institute without any experience of living abroad. The first experimental group received literature-based texts and the second one cultural-based texts that were taught by the researcher herself. The main focus of both classes was on speaking accuracy with different materials. The classes held twice a week and each session lasted for one hour and a half. Before the treatment, the learners were selected by the use of randomly convenience sampling technique in terms of their scores of Oxford Placement Test to homogenize them. The test was made of 20 multiple-choice items and three parts (reading, structure, & functional test). Those with the passing grade (15/20) or with two different scores were chosen. Due to the poor performance, 12 students were excluded from the rest of the study. After dividing them in to two classes, the students' speaking accuracy was measured via the IELTS speaking test, Cambridge ESOL, (2013) which was made of three parts (introduction, talking about a particular topic and engagement in a discussion) with the reliability of 0.83–0.86 (Bridges & Shaw, 2004) as a pretest. The voices were recorded and rated by two independent raters, the researcher and the other experienced teacher, to identify their speaking accuracy levels. To enhance the intra-rater reliability, they were scored twice by each researcher after a month.

To start the treatment, a brief introduction of extensive reading program and its benefits were provided to the students to make them aware and provide a strong rationale for attending the course. The list of the graded readers, matched with this program, was offered and they chose them on the basis of their interest and enthusiasm in the introduced genre. They were predominately from the Penguin graded readers and Oxford Bookworms from Jungle Publication. The reason for applying the aforementioned graded readers was to improve their motivation and interest in reading.

To control their reading, one graded reader was chosen every session by them. In the beginning of each session, the author and some background information on the literary text were briefly introduced to the students. Some questions regarding the theme, plot and characterization of the literary text were raised for students to discuss, which in turn could enhance students' understanding of the text. Next, they talked about the degree of their interest in the assigned story. Then they were asked to talk about their favorite character and the related parts in the story quickly. As the main point, they explained the plot of the story in their own words to the class. The researcher as a teacher talked about her points of view and enthusiasm. Simultaneously, they received feedback from the teacher. They were given time to finish their narration without any interruption which could inhibit their fluency and motivation. Six grammatical tenses were significantly focused by the teacher as major mistakes in terms of the researcher's experience as a teacher for about 20 years: past and present simple and continuous, past and present perfect as salient tenses in telling a story (Cunningham & Moor, 2005). The other grammatical points such as prepositions were considered as minor mistakes after their narrations. Due to the length of some stories like ET or Pride and Prejudice, two sessions were dedicated to them. It is worth mentioning that only 20 minutes was dedicated for the discussion part each session, so due to the allocated short period, the students' improvement was mainly related to the practice outside of the class activity.

As a posttest, the other IELTS speaking test with the same year of publication, Cambridge ESOL, (2013) was administrated to test the influence of the treatment on their speaking accuracy. The recorded voices were scored by two independent raters as well. After a month interval, they were all rated by the researcher again in terms attached IELTS scoring rubric.

5. Results

To reveal significant effect of literature and culture-based input flooding instruction on the learners' speaking accuracy, the data were entered and analyzed using the International Business MachinesCorporation (IBM) Statistical Package for the Social Sciences (SPSS) 21. Then, through running a paired sample t-test, the participants' pretest scores were compared against their posttest scores via input flooding by the use of literature and culture-based texts in order to find if any possible difference exists and to see if this type of instruction was effective. The required descriptive statistics were also computed.

5.1 Examining the Homogeneity of the Students

5.1.1 General language Proficiency

As described in the previous section, the Oxford Placement Test (OPT) was administered among the 60 participants of this study to examine their homogeneity in terms of the level of their general language proficiency. After analyzing the result of the OPT test, 48 advanced students were chosen and randomly divided into two groups (using convenience sampling technique). Twelve students were excluded as outliers. The results obtained from the test are shown in Table 1.

Because the majority of the ratios are within ±1.96 (-.007 & .67), such a result indicates that the group is homogeneous.

5.1.2 Language Speaking Proficiency

As described in the previous section, the IELTS speaking test (2013) was administered among the two groups of the participants of this study to examine their homogeneity in terms of the level of their speaking ability. To do so, the independent samples t-test was run. The results are presented in Table 2. One of the assumptions of t-test statistics is normality of the data which was checked by computing skewness and kurtosis ratios (i.e., by dividing the skewness and kurtosis values by their standard error) from the relevant values in Table 2. Because all the ratios are within ±1.96, the data could be considered normal enough; therefore, an independent sample of t-test as a parametric test could be run.

Table 2

Descriptive Statistics of the Literature and Culture Groups' Language Speaking Proficiency

The findings of Table 3 present the independent samples t-test on comparing the two groups of this study in terms of their mean language proficiency.

Evidently, the t-test results assuming the homogeneity of variances according to the Levene's test results (p>.05) show that the difference between the two groups is insignificant; t (46) = .74, p>.05 (.46), hence showing the homogeneity of the groups in terms of language proficiency.

5.2 Interrater Reliability

To assure the intra-rater reliability, the voice of the learners were recorded and scored by the researcher before and after the treatment in terms of the IELTS rating rubric. After a month interval, they were scored again by the researcher herself. To measure the interrater reliability, the learners speaking accuracy were rated by another rater who was an MA holder and IELTS examiner. The results of the inter-rater analysis was Kappa = 0.39 with p<0.001. This measure of agreement, while statistically significant, the raters have fair agreement with each other regarding scoring learners' speaking structural accuracy. As a rule of thumb values of Kappa from 0.40 to 0.59 are considered moderate, 0.60 to 0.79 substantial, and 0.80 outstanding (Landis & Koch, 1977). Most statisticians' preference for Kappa values is to be at least 0.6 and most often higher than 0.7 before claiming a good level of agreement. Therefore, there is a moderate inter-rater reliability in this study.

5.3 First Research Question

The first question focuses on the effect of the receptive input flooding through the extensive literature-oriented reading on the adult female EFL learners' speaking structural accuracy at the advanced level.

To uncover if there was a measurably huge contrast between the two methods for pretest and posttest scores in regards to input flooding instruction utilizing literature-oriented texts and its impact in light of the advanced female learners' speaking proficiency, with a significant level of .05, initial, an average score for each group's pretest scores and another average score for each group's post test scores were gotten. By the use of the International Business Machines Corporation (IBM) Statistical Package for the Social Sciences (SPSS) 21 the information analyzed utilizing. Finally, through running a paired sample t-test, the members' pretest scores were analyzed against their posttest input flooding using literature-based texts scores, to locate any conceivable contrasts and to figure out whether this kind of guideline was viable. The required descriptive statistics were also registered. The findings of the paired sample t-test are presented in Table 4.

      According to Table 4, the means of pretest and posttest regarding input flooding through literature-based texts' instruction can be compared. The pretest had a mean of 5.58 and the posttest 6.37. In addition, the standard deviation is 1.36 and 1.28 for pretest and posttests, respectively. This means that students performed better on input flooding through literature-based texts posttest, but still it cannot be claimed that they were significantly better on posttests. Hence, to check if the difference is significant paired sample t-test is used. In order to run t-test, it is necessary to check the normality of the data, which was done by computing skewness and kurtosis ratios (i.e., by dividing the skewness and Kurtosis values by their standard error) from the relevant values, the findings of which have been presented in Table 4.6. Since the majority of the ratios are within ±1.96 the data could be considered normal enough. It is only the skewness ratio of the pretest and just marginally above 1.96 (i.e., 2.05) which is could be negligible given the sample size and three other normal sets of data. The results of the paired samples t-test regarding literature-based texts instruction is illustrated in the following Table 5.

Based on the results in Table 5, it is apparent that the two-tailed probability value is .00, which is less than the alpha of .05. The result is t (23) = -7.31, p < .05, Cohen’s d = .59 medium to large effect size according to the following Cohen's threshold (Ellis, 2009).

Table 6

Cohen's Thresholds for Interpreting Effect Size

Notes. The rationale for these benchmarks can be found in Cohen (1988) at the following pages: d (p.40) and r (pp. 79-80). Supplementing Cohen's (1988) original small, medium, and large effect sizes,    Rosenthal (1996) added a classification of very large, defined as being equivalent to, greater than d=1.30 or r=.70.                            

Moreover, the reported observed value for the df = 23 is 7.31 which is greater than its critical value 2.06. Thus, the null hypothesis can be rejected and it can be stated that the results are statistically significant; hence support is given to any alternative hypothesis. That is, the null hypothesis that there is not any statistically significant difference between EFL learners' pretest and posttest scores when they receive input flooding through literature-based texts' instruction.

5.4 Second Research Question

The second question focuses on the extent of the effect of the receptive input flooding through the extensive culture-oriented reading on the adult female EFL learners' speaking structural accuracy at the advanced level.

In order to find out if there was a statistically significant difference between the means of pretest and posttest scores regarding input flooding using culture-based texts and its effect on EFL learners' speaking structural accuracy, with a significance level of .05, first, an average score for each group's pretest and post test scores were obtained. Therefore, each group of participants had one average score for pretest and posttests. Then the International Business MachinesCorporation (IBM) Statistical Package for the Social Sciences (SPSS) 21 was utilized. Finally, the participants' pretest scores were compared against their posttest scores receiving input flooding using culture-based texts through running paired sample t-test to explore any possible differences. The required descriptive statistics were also computed. The findings of the paired sample t-test are presented in Table 7.

As shown in Table 7, the means of pretest and posttest regarding input flooding through the use of culture-based texts' instruction are compared. The pretest had a mean of 5.27 and the posttest had a mean of 6.33. Also, standard deviation is .87 and .97 for pretest and posttests, respectively. This means that students performed better on posttest, but still it cannot be claimed that they were significantly better on posttests. Therefore, we have to check if the difference is significant through running paired sample t-test. In order to run t-test, it is necessary to check the normality of the data, which was done by computing skewness and kurtosis ratios (i.e., by dividing the skewness and kurtosis values by their standard error) from the relevant values presented in Table 8. Since all the ratios are within ±1.96, the data could be considered normal enough.

Considering the results indicated in Table 8, the two-tailed probability value is .00, which is less than the alpha of .05. The result is t (23) = -12.24, p < .05, Cohen's d = 1.15 large effect size. Furthermore, the reported observed value for the df =23 is 12.24 which is greater than its critical value which is 2.06. Thus, we can reject the null hypothesis in favor of the alternative one. In other words, the results are statistically significant; hence support is given to any alternative hypothesis. The null hypothesis that “there is not a statistically significant difference between the advanced EFL learners' pretest and posttest scores when they receive input flooding through culture-oriented texts' instruction is rejected. In other words, it can be stated that EFL learners' receiving input flooding through culture-oriented texts performed better in their posttests.

6. Discussion and Conclusions

The study incorporates the extensive reading program was to the language classroom based on the firm grounds that students would obtain the maximum result with the emission of substantial amounts of comprehensible input. Due to the fact that the interpretation of the results of different posttests of the findings are presented alongside the reported results, this part under the heading of discussion deals with some comments on the findings.

Regarding the first question in the study which disputed the effect of the receptive IF via the literature-oriented reading on the adult female EFL learners' speaking structural accuracy at the level of advanced learners, the result plainly maintained an affirmative reply. Thus, the first null hypothesis stating that there is not any statistically significant difference between EFL learners’ pre-test and post-test scores when they receive input flooding via literature-based texts' instruction was declined. The students represented considerable improvement in their structural accuracy. Accordingly, it supports Krashen's comprehensible input hypothesis. Cirocki, (2009) focuses on studying the practicality of implementing the extensive reading approach in a secondary-school context which ends well in a reliable result in this study. Moreover, it is worth mentioning the other study done by Turner (2012) on reading narrative literature in fiction or biographical forms, persuades social work students to deal with contexts which will help them promote such competence.

The reason for positive effect of the literature-based text can be logically depicted using reading the stories and the learners reflected their interaction with the characters, explained about the materials they could easily learn, and defined the way they would feel if they could meet them, or individuals like themselves, in the same professional contexts. Hence, by providing such equipped learning environments, learners had opportunities to develop their abilities for discovering other people's feelings and problems, and to reflect how to develop relationships across different situations.

As a consequence, utilizing literature can practically impress the learner's progress in target language. Short stories can thus be of appropriate resources to scrutinize both language and life, and enhance experiences in both. Characters in a short story world sometimes live a life like people's in the real world. They show active communication and interaction with each other and their surroundings as well as doing many things in a variety of satisfactions. Bringing literature to the class is supposed to be fun and meaningful so that the students could improve their English proficiency in mastering four skills especially their speaking ability. Ahmadian and Pashangzade (2013) in their survey highlights the positive effect of applying literature in language learning to improve the learners' reading ability, which fortify the implication of the literature in our research.

Although there have been a lot of research studies in the literature regarding the comparative examination of the implicit effects of focus on form, the present study could be considered as an additional support for implicit focus on form through input flooding. It is worth mentioning that this technique led to better accuracy levels in both groups. This effect of input flooding on speaking structural accuracy can be observed in the study by Huang and van Naerssen (1987) and Hafiz and Tudor (1990), who found that reading outside class was the most significant predictor of oral communicative ability. In another study by Mart (2012), he examined the relationship between the ER and speaking ability. The finding of this study confirmed the existence of the positive relationship since vocabulary and grammatical knowledge are two essential factors which can influence the learner's speaking performance as he claims. In the study conducted by Khatibi (2014), the findings also provided empirical support for the facilitative effect of extensive reading on speaking performance of EFL learners. This improvement of the learners in this study would be the result of the role of extensive reading in giving the learners rich background knowledge, vocabulary recognition, a high motivation for more reading, and discovery of reading strategies by learners themselves, and increasing guessing ability in context.

In regard to the second question of this paper which investigated the effect of the receptive input flooding through the extensive culture-oriented reading on the adult female EFL learners' speaking structural accuracy at the advanced level, the finding illustrated the significant improvement in their structural accuracy by the use of culture-based texts. Therefore, the second null hypothesis as no significant effect of using culture-oriented texts exists on structural speaking accuracy is declined. In the research done by Tsou (2005), culture instruction was implied within two elementary EFL classrooms for one semester to observe the effects of culture instruction on foreign language learning. As culture lessons were integrated into EFL instruction, students' language proficiency was significantly improved. Besides, they owned better interests in language learning. Su (2011), in his study tried to examine the effects of the cultural portfolio project on learners' specific aspects of developing cultural knowledge and changing the perception of native English speakers and their cultures. The finding of his study demonstrated the significance of positive effect. These mentioned findings put more emphasis on the results of this research study.

One explanation for this improvement is that using the culture-based texts helps learners move from an ethnocentric view to respect cultural differences to become more aware of diversity within culture, and find out that the media presents the surface culture of native-English-speaking countries. The next reason to support the idea is the opportunity of implementing the extensive reading programme that offers "when students talk to each other about the books they have independently read, they make a natural and often absorbing topic for conversation" (Bamford & Day, 2010, p. 93).

Using regular class discussions about students' reading thus is in close relation with the development of their speaking accuracy and English learners often find difficulty with the skills that help them string longer phrases of discourse together to acquire fluency. Students also are in the need of teacher's help in order to become an intercultural speaker or intercultural mediator who has the ability to interact with others, to find other views and perceptions of the world, to mediate different perspectives, and to be conscious of their assessing the differences. In other words, teachers should make the task in a way that facilitates learners' interaction with other small segments of another society and its culture, with the purpose of relating learners' understanding of their own cultural values with beliefs, reactions and behaviors, and inspiring them to investigate the otherness around them (Kramsch, 1998).

The quantitative findings of this research study showed and underpinned the significant effect of the input flooding on the EFL learners' structural accuracy. It can be concluded that the extensive reading program helps the students to improve their structural accuracy. Surprisingly, they have more control on their specific structures in narrating the stories (simple present/past, present/past continuous, present/past perfect).

As a result, it can be concluded that implementing the extensive reading program by the use of literature and culture-based texts can have the positive effect on the learners' speaking accuracy as well as the sense of satisfaction given to the students.

The effects observed from the provision of comprehensible input through high interest, well-illustrated story books are highly asserted by Krashen's input hypothesis (1991). In addition, the outcome of this study highlights the positive effect of simplified graded readers on the comprehensible input in terms of Day and Bamford (1998). The application of literature-based texts in the classroom and its positive effect on the learners' speaking structural accuracy is signified as the McKay's (2001) suggestion of the productive use of literature in the EFL classes.  Not only literature but also the application of the culture-oriented texts as a meaningful context and its effectiveness in language learning is emphasized in Ausubel's (1968) meaningful learning theory.

Because this study shed light on the implementation of the extensive reading program in an EFL context, it may be a useful yardstick for the incorporation of extensive reading program into language classes providing teachers with guidelines to incorporate ER into the English curriculum and use this power in their classes.

      Experience from across the world proves that ER programs would play a significant role in promoting language improvement and development. Unless the teachers firmly believe that ER is beneficial in promoting English language development among their students, they are not likely to exert their efforts to get the utmost benefits out of the program. Teachers must themselves trust this approach and try to learn it and experience it. They should make sure of that it is beneficial for both themselves and also for their students (Alami, 2009).

One of the pedagogical implications that can be drawn from this research study is that extensive reading program can be successfully implemented in EFL classes compensating for limited exposure to English language outside the classroom. In addition, teachers and educators should allocate more time and opportunities for students to read extensively by setting up this input flooding program by the use of interesting graded readers. Teachers should bear in mind that exciting materials are playing crucial roles in encouraging students to read more; as a result, the miraculous power in language acquisition should not be neglected.

As recommended by Day and Bamford (1998) and the findings of this study, it is necessary to provide students with a careful orientation to ER and guidance in their work by telling them in simple terms the rationale, purpose, and procedures of extensive reading.

Because the study unveiled positive effects of ER programs on speaking structural accuracy development, it can be concluded that this program can improve language ability, skill and confidence in learning English. In addition, the use of graded readers can pave and facilitate this way. This can be the result of the students' freedom in choosing the books they like and their effort to finish them themselves with their own pace.

It is noteworthy that the implementation of extensive reading program and enjoying the positive results are not possible overnight, but this process needs time and energy from both teachers and students which outweigh the cost. Teachers should also be a role model as a reader to make their students eager to read (Nuttall, 1996).

Narration orally without the use of the text can create a well-established connection between the teachers and the learners which will expand the structural knowledge of the learners, namely as words are heard, they will be easier to be recognized in print. The above-mentioned rapport also helps learners to internalize the data and use it productively in future opportunities.

Another point needed to be taken into consideration is the effect of implementing literature and culture-oriented texts through which the students' positive attitudes were investigated. It might also be useful and essential too, for policy makers and course designers to take the learners' responses and attitudes in consideration while designing curriculum and prescribing syllabi to them. They can think of developing textbooks in which a mixture of both native and target culture are depicted so that a balance between the two are maintained. It is also helpful for text book writers to be aware of learners' attitudes towards different entities of culture in the context of English language classroom. Therefore, the target language culture may occupy an important position in these English language classrooms. Furthermore, language teachers should be encouraged to supplement foreign language classroom instruction with discussions about the cultural perspectives of the target language society. For students with advanced language skills, presentations and discussions/debates can be used for cultural knowledge learning. For example, students can discuss the similarities and differences between cultures which can be used for skills and attitudes training to motivate students and to strengthen cultural knowledge learning.

Therefore, narrative texts (or short stories) are able to deal with the issue of authenticity as also stated by Ahmadian and Pashangzadeh (2013). Short literary stories invite L2 learners to engage in a more active and extended discussion of their involvement with the text and their personal experiences relevant to the world of the text, through which classroom interactions, learners' own experiences as important contributors to learning, and learner's orientation is covered. On the other hand, literary texts in general and short narratives in particular, serve many causes in language teaching environments; they are used mostly as an authentic window to the world of a foreign culture and society.